inc_counter looked up the per-CPU counter, incremented it in place, and then called bpf_map_update_elem to store the same value back. For a BPF_MAP_TYPE_PERCPU_HASH the pointer returned by the lookup already is this CPU's slot, so the write-back was redundant, and it is the one step that takes the hash bucket spinlock. The program runs for every received packet, not just STUN, and updates two counters each time, so a busy DERP paid two locked map updates per incoming TCP segment. Keep the update only for the miss path, which runs once per key. While here, make inc_counter and handle_counters return void. Neither caller looked at the result. TestXDP cleared the counters map before every case, so it only ever exercised the key insertion path. Add a subtest that runs the same packet repeatedly without clearing, to cover the in-place increment of an existing key. Regenerate the bpf2go outputs with clang 18.1.8 and the bpf2go from the cilium/ebpf v0.22.0 already in go.mod, which is also why the generated Go files pick up structs.HostLayout and the object name constants. Fixes #22463 Signed-off-by: Mike O'Driscoll <[email protected]> Co-authored-by: Brad Fitzpatrick <[email protected]>
DERP
This directory (and subdirectories) contain the DERP code. The server itself is
in ../cmd/derper.
DERP is a packet relay system (client and servers) where peers are addressed using WireGuard public keys instead of IP addresses.
It relays two types of packets:
-
"Disco" discovery messages (see
../disco) as the a side channel during NAT traversal. -
Encrypted WireGuard packets as the fallback of last resort when UDP is blocked or NAT traversal fails.
DERP Map
Each client receives a "DERP Map" from the coordination server describing the DERP servers the client should try to use.
The client picks its home "DERP home" based on latency. This is done to keep costs low by avoid using cloud load balancers (pricey) or anycast, which would necessarily require server-side routing between DERP regions.
Clients pick their DERP home and report it to the coordination server which shares it to all the peers in the tailnet. When a peer wants to send a packet and it doesn't already have a WireGuard session open, it sends disco messages (some direct, and some over DERP), trying to do the NAT traversal. The client will make connections to multiple DERP regions as needed. Only the DERP home region connection needs to be alive forever.
DERP Regions
Tailscale runs 1 or more DERP nodes (instances of cmd/derper) in various
geographic regions to make sure users have low latency to their DERP home.
Regions generally have multiple nodes per region "meshed" (routing to each
other) together for redundancy: it allows for cloud failures or upgrades without
kicking users out to a higher latency region. Instead, clients will reconnect to
the next node in the region. Each node in the region is required to be meshed
with every other node in the region and forward packets to the other nodes in
the region. Packets are forwarded only one hop within the region. There is no
routing between regions. The assumption is that the mesh TCP connections are
over a VPC that's very fast, low latency, and not charged per byte. The
coordination server assigns the list of nodes in a region as a function of the
tailnet, so all nodes within a tailnet should generally be on the same node and
not require forwarding. Only after a failure do clients of a particular tailnet
get split between nodes in a region and require inter-node forwarding. But over
time it balances back out. There's also an admin-only DERP frame type to force
close the TCP connection of a particular client to force them to reconnect to
their primary if the operator wants to force things to balance out sooner.
(Using the (*derphttp.Client).ClosePeer method, as used by Tailscale's
internal rarely-used cmd/derpprune maintenance tool)
We generally run a minimum of three nodes in a region not for quorum reasons (there's no voting) but just because two is too uncomfortably few for cascading failure reasons: if you're running two nodes at 51% load (CPU, memory, etc) and then one fails, that makes the second one fail. With three or more nodes, you can run each node a bit hotter.